One of the most frequent questions I get asked is whether wearing masks really helps prevent the transmission of COVID-19. I answer that the data is not entirely conclusive, but clearly this is a respiratory infection, and it stands to reason that lowering aerosol levels should decrease infection rates.
There have been small scale studies in hospitals that have suggested mask wearing was important, but no large controlled trial of the general population has conclusively shown efficacy. One of the big problems is the difficulty of doing a group-controlled trial on the effect of wearing a mask. Most people who wear a mask take additional steps such as social distancing to reduce their likelihood of being infected. Social distancing is hard to control for in a group not wearing masks, and those people are often less likely to take these additional steps.
These challenges are very apparent in a study published today in the “Annals of Internal Medicine.” The study was conducted in Denmark in April and May of this year and included almost 5,000 adults that spent more than three hours per day outside their home. The subjects in the study were either asked to follow social distancing rules and wear a mask or follow social distancing rules without wearing a mask. The organizers of the study encouraged the use of masks and provided the mask group with a supply of 50 surgical masks, but there was no active enforcement of mask wearing.
Remarkably, at one-month infections with SARS-CoV-2 were seen in 1.8% of the mask cohort and in 2.1% of the control group. This was not statistically different as the 95% confidence intervals range from a 46% reduction in infections to a 23% increase. The were half as many instances of clinical illness in the mask vs. the control group (5 vs 10). Regardless, this trial obviously did not conclusively validate the use of wearing a mask, although it trended toward fewer infections in the mask group.
While this would seem to be a negative study, there were a number of problems with the work. The results included significant missing data points as a number of the participants dropped out. The participants were obviously not blinded (and potentially biased) to the intervention, and the reports of COVID-19 infection were self-reported in many cases by the participants. The rate of infection in Denmark was also low at that point in time, further reducing the power of the study.
Rather than disprove the utility of masks, this study points out the difficulty in trying to test their usefulness. One can only imagine that testing the effectiveness of masks is as hard as it seems to be to use one correctly!